Genetic factors play a crucial role in the development of non-insulin-dependent diabetes mellitus (NIDDM). However, efforts to identify NIDDM genes have not been successful because of NIDDM's complexity and heterogeneity. As a new strategy for finding NIDDM genes, this project will investigate early-onset, autosomal dominant NIDDM - a subtype of diabetes that is more strongly determined by genetic factors than common NIDDM but, like common NIDDM, is characterized by the presence of insulin-resistance. Using a large panel of informative families manifesting this type of diabetes, NIDDM genes will be searched by means of two complementary approaches: 1. genome-wide screening, and 2. studies of candidate genes. DNA and clinical information are already available at the Joslin Diabetes Center for most of these families. A 10 cM genome screen for linkage with NIDDM will be performed using a novel DNA pooling technique which requires a smaller genotyping effort than traditional methods. Linkage with NIDDM will be also evaluated for candidate genes coding for both known and newly identified proteins involved in insulin action (IRS-2, IRS-3, IRS-4, PI-3-kinase, SHP-2, Grb-2, Shc, GRB-1Rbeta, and several other newly identified SH2 proteins expressed in skeletal muscle). Linkage with NIDDM will be evaluated by both parametric and non-parametric methods. Genes located in the chromosomal regions identified by the genome screen, as well as candidate genes found to be linked with NIDDM, will be screened for mutations. The investigators note that identification of genes that are responsible for the development of early-onset NIDDM will have a profound impact on research on insulin action that are involved in the development of insulin-resistance, with clear implications for studies of the etiology of NIDDM and the development of drugs to treat insulin resistance.